Peak Watch

Living On the Edge: OPEC Spare Capacity

The Times They Are A-Changing'
     — Bob Dylan

It's time for some straight talk on the oil supply. On September 12th the Organization of Petroleum Exporting Countries (OPEC) announced that they would raise production 522,000 barrels per day (b/d) to alleviate demand-side pressures in the 4th quarter of this year. Few people in the oil markets were impressed by this conciliatory gesture, for the Nymex price for light, sweet crude closed at $80.09 a barrel on Thursday, September 14th. Let's examine the oil market conditions driving the rising oil price. Demand is strong and slated to increase as winter approaches. On the supply-side, the most important factor is how much surplus production capacity OPEC can bring to bear to ease prices should they choose to do so.

Short-term price movements are insignificant¹ because they are influenced by ephemeral events such as profit-taking and reactions to weekly inventories. The longer term price trend is significant (graph from tradingcharts.com, left) because it tells us about the evolving supply & demand balance. The market has been increasingly out of balance for more than 7 years now, as the price signal indicates. The sharp dip in prices that took place last fall/winter was an anomaly in a volatile market when viewed from a longer term perspective. The price trend has resumed its up-and-down but inexorable rise since the recent low point in January, 2007—assuming all things remain equal on the demand-side of the market (i.e. no recession).

You will hear many reasons offered to explain why the oil price is high—speculators entering a futures market in backwardation, the weak dollar, intermittent hurricane threats, lack of refining capacity, disruptions in Nigeria and Iraq, geopolitical tensions with Iran, etc. Don't believe these stories. OPEC representatives will tell you that the oil markets are "well balanced." Don't believe that either. On the supply side, OPEC's spare capacity is the paramount issue, for if there isn't much of it, a tight market is highly vulnerable to any unplanned disruption. Surplus capacity is supposed to be a buffer that can be tapped when demand outstrips supply, but if the world were swimming in readily available high-quality oil, the price would not be fluctuating around $80/barrel. The record nominal price demonstrates market fundamentals at work. 

The issue of OPEC spare capacity is contentious because nobody knows what it is. Here's what MarketWatch reported on September 12th:

What is remarkable about this latest spike in oil prices is not the usual knee-jerk reaction, but the fact that it comes on the heels of the Organization of Petroleum Exporting Countries' pledge to raise output to take some of the heat off prices.

Energy traders aren't buying it...

No, the skepticism has ... to do with production capacity or lack thereof. The guys on the Nymex trading floor are questioning whether OPEC has the ability to lift output by another 500,000 barrels a day, the number they agreed on at their meeting Tuesday in Vienna.

Nymex traders don't think OPEC can produce the goods. Contrast this with OPEC Spare Capacity Keeps On Rising from Petroleum Intelligence Weekly published on September 3rd (graph left). Capacity reaches a whopping 5.226 million b/d at the end of 2008. The reliable current capacity—€”not counting Nigeria—€”is listed as 3.638 million barrels at year-end 2007. These numbers also reflect the view of OPEC propagandist Dr. Nimat B. Abu Al-Soof, who claimed that there will be "39.7 million b/d of crude capacity by the end of 2010, representing an increase of about 5 million b/d from end 2006" (excerpted from his speech The Role of OPEC Spare Capacity at the Offshore Technology Conference in May, 2007). The estimable Dr. Al-Soof's assessment, as evaluated against the IEA's August figure for OPEC production, assumes an OPEC spare capacity of 4.3 million barrels per day at the end of 2006 (see below).

In a world that is utterly dependent on the oil supply, such divergent opinions shouldn't exist. There is no excuse for it. Nymex traders wonder whether OPEC can even raise production half a million barrels per day, whereas Petroleum Intelligence Weekly confidently states that they could raise production by well over 3 million barrels per day. Let's sort this out, setting some realistic expectations about near-term supply availability and prices.

A Quick Primer

Conditions (1) through (3) describe the supply-side of the world oil market. Equation (1) is taken from Ferdinand Banks' Economic Theory and OPEC, Part 2. Skip this part if you know all this stuff.

1. dQ_O = dQ_M -€“ dQ_N, where Q_O = the demand for OPEC oil, Q_M is the market demand, and Q_N is non-OPEC supply.
2. dQ_S = dQ_C - dQ_O, where Q_S = OPEC spare capacity, Q_C is total OPEC production capacity and Q_O is as defined in equation (1). Spare capacity refers to the "maximum sustainable production capacity, defined as the maximum amount of production that: 1) could be brought online within a period of 30 days; and 2) sustained for at least 90 days" (EIA). Capacity also refers to crude + condensate production only. Natural gas liquids, which are not subject to quota restrictions, are put on the non-OPEC side of the ledger.
   
3. Non-OPEC spare capacity is always effectively = 0, except for maintenance or other temporary disruptions.

The (satisfied) demand for OPEC oil (equation 1) is taken as equal to the cartel's production, which was 30.4 million barrels of day (crude + condensate) in August of 2007 according to the IEA's Oil Market Report for September. Useful total OPEC capacity (and therefore the spare capacity) is subject to downstream market conditions as described in the next section.

All of the terms in the equations (1) and (2) are in dispute. The dispute about the change in demand on OPEC oil is called The Call on OPEC. This quantity (in barrels per day) is calculated differently by OPEC and the IEA, with the cartel low-balling the estimated demand and the IEA exaggerating growth projections. Similarly, OPEC overestimates production outside the cartel whereas the IEA increasingly makes a more pessimistic, albeit realistic, assessment. The satisfied demand for OPEC crude affects equation (2), which defines OPEC's spare capacity. Non-OPEC production has topped out for this year—what you see is the most you're going to get (ASPO-USA, September 12th). The IEA's August OPEC production number is the baseline for this analysis.

What is OPEC's Spare Capacity?

The EIA made an historic revision in their assessment of OPEC spare capacity on September 11th, the day of the OPEC meeting. Surplus capacity is estimated to be 2.41 million b/d in the 2nd quarter of 2007, with Saudi Arabia accounting for 79% of that. The former estimate gave a range from 1.90 to 2.40 million b/d for the Kingdom, but the EIA has now settled on the low-end number. The energy agency raised their estimate for the other OPEC members to 0.52 from 0.37 million b/d. Only Kuwait (200,000 barrels) and the UAE (100,000 barrels) are able to increase flows ≥ 0.10 million barrels per day. Algeria can provide 60,000 barrels per day, Iran 50,000, and Venezuela 30,000.

It is useful to make further distinctions about surplus capacity. Demonstrated capacity refers to the difference between a country's highest production level since the start of 2005 and its average monthly output through May, 2007 as recorded by the EIA. The country's assumed capacity is any estimate of capacity beyond that which has been demonstrated. Saudi Arabia's production high water mark was 9.60 million barrels per day, last seen in September of 2005. Therefore, that number is the Kingdom's demonstrated capacity, and the EIA now assumes that the Saudis can produce an additional 0.90 million b/d if called upon. Other OPEC members provide the rest, a potential 0.52 million barrels daily.

In late September of 2005, Reuters ran the story Saudi Arabia finds no takers for additional oil. At the time, most Gulf of Mexico production was shut-in from the hurricanes and the oil price had surpassed $70/barrel for the first time. The world was in a panic. Here are some excerpts—

There are "no takers" for Saudi Arabia's offer to pump more oil for a global market pressured by soaring prices and refining constraints, the kingdom's oil minister [and de facto OPEC head] Ali Al Naimi said yesterday [on September 28th, 2005]...

Al-Naimi also told a World Petroleum Congress in Johannesburg the kingdom would stick to plans to gradually boost its total oil output capacity to about 15 million barrels per day from about 11 million at present, keeping ample spare available.

But he emphasized that Saudi Arabia had not found customers for its existing excess capacity [which is] the lion's share of OPEC's offer last month to pump up to two million bpd more oil, filling any supply shortages caused by hurricanes in the US.

"There are no takers," al-Naimi told delegates after a speech in which he reassured markets on the ability of global suppliers to meet rising demand for oil over the next 3-4 years.

It is now almost exactly two years later. Oil is trading at $80.90 a barrel (at this moment) and the U.S. Department of Energy believes that Saudi surplus capacity was only 1.90 million barrels per day in the 2nd quarter of 2007. The EIA gives a lower preliminary estimate of 2.17 million b/d for OPEC as a whole in the 3rd quarter. But let us take al-Naimi at this word, at least about there being "no-takers" for his oil back in 2005. Why were there no buyers for Saudi Arabia's oil? The glass-is-more-than-half-full optimists writing reports at Petroleum Intelligence Weekly provide a clue.

The quality of OPEC spare capacity clearly matters, as was seen at the end of 2004, when Saudi Arabia had to dip into its heavy [and sour] offshore and Manifa fields to meet rapidly growing Asian demand. Sweet-sour differentials ballooned as a result...

The most plausible hypothesis is—and always has been—that Saudi Arabia could not sell any more oil to the market in the fall of 2005 because there was no more demand for its heavier sour (sulfur-laden) crude—the market was saturated. The Kingdom has cut production 1 million barrels per day since then, with almost all of the decrease consisting of this same "dirty" crude formerly bound for Asian markets (Bloomberg, March 12, 2007). What else has changed? Other than 300,000 light, sweet barrels per day from Haradh III extension—the Saudis are quite proud of it—put on-stream in early 2006, there have been no capacity increases. The Saudi Aramco 2006 annual report also admits that production in the Kingdom fell off 1.7% in that year (0.153 million barrels).

It is uncertain exactly how much of Saudi Arabia's current production for sale is light, sweet oil as opposed to their Arab Heavy and Arab Medium sour grades with API gravity values below 30°. According to the IEA's World Energy Outlook 2005, Saudi Arabia's capacity was 10.352 million barrels in 2004—The Oil Drum provided a breakdown at the time. Of this capacity, 23% came from the heavier, sour fields Safaniyah, Manifa, Zuluf and Marjan. Adding in Haradh, Saudi Arabia has about 8 million barrels of light, sweet crude production capacity. In all likelihood, any rise in the Kingdom's production will consist almost entirely of low quality crude along with marginal amounts of light, sweet oil.

Although there are signs of change in Asian refining to expand processing of sour crudes, the downstream market has not changed appreciably since 2005. The only reasonable conclusion is that Saudi Arabia's demonstrated capacity of 9.60 million b/d is also their effective capacity (give or take a hundred thousand barrels or so), at least as far as the downstream market is concerned. Their surplus capacity is therefore about 1 million barrels. Any assumed capacity beyond that—the EIA estimates 0.90 million barrels—does not affect the world oil balance.

It is thus no wonder that oil prices have gone up since OPEC announced they would raise actual (not quota level) production by 522,000 barrels per day effective November 1st. According to Platts' OPEC says Saudis to assume almost two-thirds of output hike, Saudi Arabia will raise production 327 thousand barrels per day starting in November, 2007. This will most likely be heavy, sour crude sold back to the Asian markets. Smaller increases will come from Kuwait (85,000) and, astonishingly, Venezuela (112,000), which opposed the production hike. Iran's "new output target from November (-52,000) is lower than Iran's estimated production in August." The OPEC increase will be completely offset in November by scheduled maintenance and upgrades at the UAE's Zakum and Umm Shaif fields starting in late October (Reuters). The world awaits the results of OPEC's largess.

When al-Al Naimi "reassured markets [about] the ability of global suppliers [like Saudi Arabia] to meet rising demand for oil over the next 3-4 years" in 2005, he was spouting nonsense. OPEC has taken crude oil off the market, and that has partly driven the price rise. Now they can't put enough high quality oil back to restore the supply & demand balance. If OPEC is the steward of the world's future oil supply, we're all in a whole lot of trouble. Destructive price spikes are a large risk because the surplus capacity we all depend on isn't there to bail us out. The world is now living on the edge, and oil consumers would be well-advised to change their behavior before events do it for them.

A future column will deal with OPEC's future spare capacity and supply-side market conditions.

Contact the author at [email protected]

1. For some perspective—and your amusement—I quote the BBC's story (February 15, 2000) Oil reaches $30 a barrel with the accompanying graph labeled "soaring oil prices".

The price of oil has surged to its highest level since the 1991 Gulf War. One barrel of crude oil is now worth $30.10. Oil futures rose as high as $30.40 before falling back slightly in trading in London...

US President Bill Clinton said the rise was "deeply troubling" and refused to rule out any US action to deal with the situation.

If Hilary is elected president, perhaps Bill can again look into—or advise his wife on—this "deeply troubling" situation. Apparently, he has read Richard Heinberg's book The Party's Over.

 

A Non-OPEC Progress Report

It doesn't work to leap a twenty-foot chasm in two ten-foot jumps.
     — American Proverb

A data-driven analysis of new oil projects outside of the Organization of the Petroleum Exporting Countries (OPEC) reveals that production growth is now surpassing the dismal performance exhibited in 2005 and 2006. New project schedules and the EIA's supply data make it possible to take a close look at the non-OPEC supply situation, and make a rough estimate of how 2007 will turn out. Many developments are scheduled to come on-stream in the 2007-2009 period but new projects peter out thereafter. It's important to keep our eye on the ball by examining how declines or delays affect overall non-OPEC production growth during these crucial years.

A little noticed news item appeared in Rigzone on September 5, 2007—Petrobras Lowers 2007 Domestic Output Targets.

Brazil's federal energy company Petrobras (NYSE: PBR) has reduced its 2007 production target from 1.9 Mb/d to 1.84 Mb/d, company E&P director Guilherme Estrella told reporters. [Mb/d = million barrels per day]

"We had a first half full of operational problems that will not happen again. We had problems in the ramp-up of oil platforms P-34, P-43, P-48 and P-50," Estrella said. Problems included fires and project postponements...

Petrobras forecasts average domestic oil production of 2.05 Mb/d next year... Petrobras ended 2006 with average domestic oil production of 1.91 Mb/d.

Brazil's national oil company Petróleo Brasileiro S.A. (Petrobras), is a superstar in the non-OPEC world. (See the Wall Street Journal report How a Sleepy Oil Giant Became a World Player, published on August 30, 2007.) Petrobras is a leader in offshore technology, and Brazil is one of the few countries outside of OPEC that can actually raise yearly production. It comes as a surprise that Petrobras' domestic production target is lower for 2007 than it was in 2006. This can't be good news.

The non-OPEC supply situation going forward is a major concern. OPEC¹ has cut its output, placing the burden squarely on the rest of the world to keep up with increased demand. Growth outside the cartel has barely outpaced production declines accompanying depletion in the existing production base. This problem is illustrated by a modified version of Chris Skrebowski's Oil-a-Gator, first introduced in The Power of Declines (graph left from ASPO-USA, April 4, 2007). Global output must fall if declines overwhelm the flows of new oil coming on-stream over time—regardless of the reasons why output has fallen.

OPEC's cuts have exceeded the rest of world's ability to satisfy growing demand, so yearly world oil production has been gradually declining since the beginning of 2005. OECD countries are currently drawing down inventories (EIA) to make up the difference. According to the EIA data, oil (crude + condensate + natural gas liquids) production was 81.430 million barrels per day in 2005 and 81.321 million in 2006. Production stands at 81.209 million barrels per day through May of 2007 (five-month average). OPEC's output has fallen 1.190 million b/d during these 29 months. Global oil production is down 221 thousand b/d from the historical peak in 2005. The numbers indicate anemic growth in non-OPEC supply resulting in a net addition of 0.969 million b/d during the period, an average increase of 33,400 barrels (.0065%) per month.

The IEA's Medium-Term Oil Market Report sets the decline rate outside of OPEC at 4.6%. Skrebowski's Megaprojects database, which excludes fields producing less than 40,000 b/d, indicates that 2.905 million b/d was scheduled to come on-stream from non-OPEC countries during 2005 and 2006. Declines in existing production or project delays took away the rest.

An Oil & Gas Journal special report Upstream production capacities to advance in many countries "lists numerous announced major upstream projects in 50 countries" along with scheduled target dates and peak production flows (July 23, 2007). The list provided by author Guntis Moritis allows us to see how non-OPEC oil production is progressing. Table 1, shown after the text below, evaluates every new liquids development ≥ 20 thousand b/d listed by Moritis for 2007 except Sakhalin I, which came on-stream in 2005 and peaked in February, 2007 at ≅ 250 thousand b/d. This analysis is necessarily preliminary, but permits a reasonable rough estimate of how this year will turn out.

2007 Non-OPEC Supply Analysis

Non-OPEC production is up 0.638 million b/d over 2006 through May of 2007 according the EIA data. This is an impressive number, representing about 66% of all the growth since the beginning of 2005. The Table 1 data indicates that this net addition may be close to the ceiling for non-OPEC additions in 2007, which appears to be in the vicinity of 700 thousand b/d.

A note on assessments: The graph (left) shows a typical field production profile. Production will ramp-up, achieve peak flows and then decline monotonically (in this example). The start-up may be extended, peak flows may last longer or not be achieved at all, or exceeded, etc. Declines may not be monotonic over time (e.g. exponential at a constant rate) depending on applications of oil recovery technologies. All of this depends on the geology of the reservoir and many other factors. In his Oil & Gas Journal article, Moritis states that "if all the projects listed had a peak production rate in the same year, the total world production capacity would increase by 30 million b/d" for the entire list. Projects peak at different times of course and do not always arrive on schedule. The total 2007 non-OPEC addition of 2.409 million barrels per day from new field developments must be viewed in this light. Projects that were delayed or did not achieve peak flows in previous years are added to this year's tally, while production from many new 2007 projects will be pushed into subsequent years for the same reasons. The oil supply is a moving target.

Table 1 indicates that roughly 780 thousand barrels per day have come on-stream during 2007 as of September 1st, with about 1.63 million b/d scheduled for later this year or delayed into subsequent years. Here are some of the details. (Figures cited are compared to the year-end 2006 number.)

1. Russia's output has risen 182 thousand b/d, the 2nd largest increase among non-OPEC countries this year. In so far as Russia has no new projects in 2007, most of this growth must have come from Sakhalin and some growth in its existing production base. See For Russian, An End to Growth is In Sight (ASPO-USA, August 15, 2007).
2. Buzzard has experienced minor difficulties, but is operating at about full capacity (look at Table 1). The UK's production is up 78 thousand b/d. The rest of the 200 thousand b/d from Buzzard is   offset by declines elsewhere.
3. Azerbaijan is up 201 thousand barrels per day, the largest increase among non-OPEC countries. Most of this must be coming from the ACG Phase II expansion at East Azeri (look at Table 1).
4. Canada's production has increased 127 thousand b/d this year, with some of that coming from the tar sands. It appears that conventional production from existing fields in Atlantic Canada has contributed much of the increase.
5. There is no indication that the Gulf of Mexico developments, Atlantis and Neptune, will ramp-up this year or contribute significant quantities of oil if they do. The same remarks apply to Alvheim (Norway). Some projects such as Puffin (Australia) or Piranema (Brazil) will probably contribute small oil flows in the 4th quarter.
6. Malaysia's production is slightly down, but the current EIA data does not reflect Kikeh, which was successfully put into production in August. Production at the field will peak somtime in 2008 and does not add much to this year's total.
   
7. Brazil has already stated (link op. cit.) that they expect a small daily production decrease this year, and expect to average 2.05 million b/d in 2008, an increment of only 21 thousand barrels. Golfinho and Roncador (P-52) are set to come on-stream in October, but obviously won't affect this year's non-OPEC production hike much.

These considerations point to an increase of at most 700 thousand b/d in non-OPEC supply for 2007, a substantial improvement over the 2 previous years. Growth beyond this ceiling requires that new oil or existing field upgrades must surpass non-OPEC declines. If the IEA's estimate of the decline rate for non-OPEC production is accurate, the world must put 2.324 million b/d on-stream each year to offset depleting fields. This number is higher than the peak flows total in Table 1. It seems clear that new oil from projects scheduled in previous years is making up much of the difference. With so many 2007 projects experiencing delays, declines may offset additions for the rest of the year, thus lowering the non-OPEC increment. Updates to the EIA data will allow us to follow the situation more closely.

Production growth is stronger than it was in the previous two years, but that doesn't mean that the non-OPEC supply picture is rosy for the indefinite future. New projects become scarce after 2009 according to both Skrebowski's and Moritis' schedules for upcoming upstream projects. These schedules are reasonably accurate because of the long-lead times required to get new oil into production. Analysts like the IEA and the Centre for Global Energy Studies (CGES) have been forced to revise their non-OPEC forecasts downward in recent years (ASPO-USA, April 11, 2007). We await the final verdict for non-OPEC supply, which won't be in until the 1st quarter of 2008.

Table 1 (with links to project reports)

Project	Peak Flow	Country	Notes
Polvo	50	Brazil	on-stream, Devon, first foreign operator
Espadarte Sul	100	Brazil	on-stream, first oil in January, 2007; Petrobras (and others below)
Golfinho	100	Brazil	first oil in October, 2007; Cidade de Vitória FPSO
Roncador P-52	180	Brazil	first oil in October, 2007; peak in 2nd half of 2008
Roncador P-54	180	Brazil	delayed, inaugurated FPSO P-54 rig in August
Piranema	30	Brazil	first oil in September, 2007
Puffin	40	Australia	FPSO in place, first oil in late September, 2007
ACG Phase 2	260	Azerbaijan	Azeri-Chirag-Gunashli, East Azeri; part done and being completed now
Peng Lai	72	China	Phase 2; delayed to 2009; ConocoPhillips in tax dispute with CNOOC
Xijiang 23-1	100	China	target was late 2007, no news after 2005
Wenchang	120	China	same as Xijiang, will be produced in "coming years" — CNOOC
Okume	60	Equatorial Guinea	first oil late 2006, peak in 2008
Mangala	80	India	delayed until 2009, requires a pipeline
Kikeh	120	Malaysia	on-stream in August, 2007, peak in 2008
Tui	50	New Zealand	on-stream in August, 2007, peak in 2007
Alvheim	85	Norway	delayed until 4th quarter, FPSO not ready
Ormen Lange	35	Norway	gas liquids, commercial gas production in 4th quarter 2007; peak gas flows in mid-2008
Statfjord	70	Norway	"late life project", compression trains requiring upgrades delivered in September, 2007, becoming a gas field with associated oil
Buzzard	200	United Kingdom	first oil in January, 2007; some recent troubles but outlook is good
Chestnut	30	United Kingdom	production platform being readied for first oil late 2007
Neptune	50	United States	ultra-deepwater, delayed, BHP Billiton was still drilling in April; costs up 35%
Atlantis	200	United States	BHP Billiton still expects first oil in 2007; little information; costs up 9%
Ca Ngu Vanf	70	Vietnam	means "golden tuna"; was just getting underway in 2006 with first oil "expected" in 2007; unknown
Surmont	25	Canada	oil sands SAGD, phase 1, delayed; 300 thousand b/d expected "within the next decade"
Cold Lake	30	Canada	oil sands (Imperial oil) expansion, unknown
Long Lake	72	Canada	oil sands, huge cost overruns, cost inflation, first oil delayed to mid-2008, peak 12-24 months afterwards

1. This story is not affected by the announced OPEC production increase of 0.5 million b/d. I will write about that in next week's column.

Contact the author at [email protected]

PetroProzac, Dasani-Style

I was on Prozac for a long time. It may have helped me out of a jam for a little bit, but people stay on it forever. I had to get off at a certain point because I realized that, you know, everything's just OK.
     — comedian Jim Carrey

When the Wall Street Journal published Message in the Drink Bottle: Recycle on August 30, 2007, the words "petroleum", "oil" and "natural gas" did not appear in the text. Branded-Water Craze Adds to Piles of Plastic, Spurring Innovation at Coke, Pepsi, the sub-headline read. Peak oil is usually thought of as a liquid fuels crisis, but all of the other uses mankind has found for hydrocarbons should not be left out of the equation. As with gasoline, we consume plastic water bottles (pictured below left) as if there were no tomorrow.

The plastic in water bottles is polyethylene terephthalate (PET or PETE). Where does this magical substance come from?

PET is a polymer, a substance consisting of a chain of repeating organic molecules with great molecular weight. Like most plastics, PET is ultimately derived from petroleum hydrocarbons. It is created by a reaction between terephthalic acid (C₈H₆0₄) and ethylene glycol (C₂H₆0₂).

Terephthalic acid is an acid formed by the oxidation of para-xylene (C₈H₁₀), an aromatic hydrocarbon, using just air or nitric acid. Para-xylene is derived from coal tar and petroleum using fractional distillation, a process that utilizes the different boiling points of compounds to cause them to "fall out" at different points of the process.

Ethylene glycol is derived from ethylene (C₂H₄) indirectly through ethylene oxide (C₂H₄0), a substance also found in antifreeze. Ethylene is a gaseous hydrocarbon that is present in petroleum and natural gas, but is usually derived industrially by heating ethane or an ethane-propane mixture.

Ethane and propane are common components¹ of natural gas liquids, and serve as the primary feedstock to ethylene(-propylene) crackers, which accounts for this part of PET. Naphtha, which is produced during crude oil refining, is a common feedstock for an aromatics complex that derives the para-xylene. The"petro" in petrochemicals is there for a reason.

What is the energy cost to produce a plastic PET bottle? This is a hard question to answer because boundaries must be put on the calculation. The Pacific Institute cites a PlasticsEurope study (click on PET Bottles in this graph to download it) to make a rough calculation which we will follow here. According to the Wall Street Journal, a half-liter Dasani (Coca Cola) bottle now weighs 16.0 grams, a reduction of 5 grams off the older weight. The study indicates that it requires 103.94 megajoules (MJ) to produce 1 kilogram of PET bottles. A simple calculation reveals that it takes 3.326 MJ to produce a one liter Dasani bottle weighing 16 grams, excluding the cap and packaging. This is equivalent to 0.00052 barrels of fuel oil. According to the Beverage Marketing Corporation, Americans consumed 31.2 billion liters of bottled water in 2006. If all of this water had been Dasani served up in PET bottles, the oil equivalent would be 16.224 million barrels. This amounts to about three-quarters of one day's oil consumption in the United States.

PlasticsEurope extols the virtues of plastics. The library page is especially illuminating. "Plastics are fantastic ... today's lifestyles simply wouldn't be possible without plastics." One can't help but agree. People in the 19th and most of the 20th century got by without them somehow, but they didn't have a computer, nifty bicycles, etc. Under Facts and Figures, we learn "that plastics consume just 4% of the annual consumption of oil and gas, yet they are one of the most resource-efficient materials around." No argument. It is the Resource Efficiency section that matters—

Plastics are much too valuable to throw away, and putting plastics waste into landfill is like throwing oil away! PlasticsEurope supports eco-efficient solutions such as the diversion of ‘calorie rich’ plastics waste from landfill and end-of-life management of this material through recycling or recovery.

Water, Water Everywhere, a report from the Container Recycling Institute (CRI), describes the growth of non-carbonated beverage consumption in the United States and its attendant problems. CRI estimates that "approximately 18 million barrels of crude oil equivalent were consumed in 2005 to replace the 2 million tons of PET bottles that were wasted instead of recycled."

The provenance of the CRI number is unknown², but a precise calculation can be carried out using CRI's data source for their graph (left) and the conversion factors introduced above. The National Association for PET Container Resources indicates that 3.905 billion pounds of PET plastic bottles available on store shelves were thrown out in 2005, as opposed to the 1.170 billion pounds that were recycled. The total includes all PET bottles, e.g. for carbonated beverages, not just water. Doing the conversions, it appears that 95.735 million barrels of fuel oil equivalent was "thrown away" when these PET bottles were put in the garbage, which is a little less than 5 days of United States consumption. PET water bottles make up 17% of the total. (This calculation does not include possible energy recovery—see below.)

After the PET bottle is manufactured, energy is expended for the cap, the label and packaging. The bottles must be filled with filtered water, perhaps with some flavoring and minerals added, and transported to a retail destination where someone buys it and drinks it. At that point, "the empty PET packaging is discarded by the consumer after use and becomes PET waste. In the recycling industry, this is referred to as post-consumer PET." If the consumer recycles it, the plastic is subject to mechanical recycling—

Once collected, the plastics are delivered to a material recovery facility (MRF) or handler for sorting into single resin streams to increase product value. The sorted plastics are then baled to reduce shipping costs to reclaimers.

Reclamation is the next step where the plastics are chopped into flakes, washed  to remove contaminants and sold to end users to manufacture new products such as bottles [e.g. more PET water bottles], containers, clothing, carpet, plastic, etc.

If the PET bottle is thrown into the garbage, it either ends up in a landfill, or is converted to heat and electricity in a waste-to-energy (WtE) process, usually through incineration. PET plastic is composed of complex hydrocarbons, and thus has a high calorific value. Efficiency in WtE processes vary, but must be improved. This is called "energy recovery" in the post-consumer disposal world.

We have followed the life of the PET water bottle from the cradle to the grave. In the best case, some small fraction of the total energy expended will be recovered at the end via WtE. Otherwise, the plastic will be recycled or in the worst case, the entire process is a dead loss. All of this energy was used up so that a thirsty "consumer" could drink bottled water instead of tap water, and in the case of Dasani, for the profitability of the Coca Cola Company and any non-subsidiary corporate entities participating in the PET bottle life cycle. All of this economic activity boosts growth which adds to various countries' Gross Domestic Products, provides jobs and gives economists a warm fuzzy feeling. That's the plus side.

On the negative side of the ledger, much of the energy expended to make PET bottles is just thrown away, especially in the United States. (Energy recovery and recycling is far more prevalent in Europe than it is in the U.S.) The author has not been able to find a study that calculates the net energy loss—up to the point of reuse in the recycling case—for each possible outcome over the entire PET life cycle. Garbage dumps continue to fill up, but a lot of the PET plastic ends up in the oceans—see Polymers Are Forever (Orion Magazine, May/June 2007). The North Pacific gyre is a troublesome destination. The plastic, which breaks down into durable tiny particles a few millimeters or less in size, is ingested by open ocean filter feeders, which eventually kills them. There are the usual carbon dioxide emissions associated with making PET bottles, which contributes to global warming.

Under pressure from municipalities and environmentalists, the bottled water manufacturers are reducing the amount PET in their product and acknowledging their obligation to encourage recycling according to the Wall Street Journal (graphic left, link op. cit.). The issues presented are clogged landfills and carbon dioxide emissions that contribute to global warming. The bottling companies are only now "starting to warm up to financial incentives for recycling", which they have consistently opposed in the past. Eleven states, including California and New York, have deposit laws. Like the Coca Cola Company, Pepsi—Aquafina, which is also filtered tap water—has reduced the weight of their bottles but only 10% of them are made from recycled PET.

Does the world need PET water bottles? No. Many of us were born during a time when they didn't exist, but we managed to muddle through somehow. Why do "consumers" think they need them? There is a perception "that municipal water is not as safe as it could be" and sometimes that's true, but "New York City’s water is every bit as good as bottled Evian" according to Walt Boyes, principal of Spitzer and Boyes, consultants in water treatment and distribution. This is not the real reason people drink bottled water, however. It is marketing that drives the consumption, and consumption is growing briskly (graph left, from CRI).

“Branding is extremely important for water,” says Chiranjeev Kohil, professor of marketing at California State University at Fullerton. “In a lot of categories, you can duplicate products and get an edge on quality or attributes, but that edge can be shaved off very quickly by competitors. In the water category, there is no technological superiority. The only thing that differentiates one water from the next is the brand.”

As with wanton³ consumption of liquid transportation fuels, it does not seem to have occurred anyone that throwing oil away is a more fundamental problem than clogged waste dumps or recycling efforts. It is the availability of tap water filtered by reverse osmosis delivered in a PET bottle that's at issue in the case of Pepsi or Coca-Cola. Exotic products like Fiji Water have even higher energy costs than those cited here. Petrochemicals are taken for granted. Our society is taking PetroProzac again. Recycling is better than not recycling, and lighter bottles reduce the volumes of fossil fuels feedstock required. The real problem, however, is the way the world does business. We need to stop all this foolishness now.

1. Refining and petrochemical feedstocks are obviously very complex subjects. If you want to learn more, there are many excellent books, e.g. Petroleum Refining: Technology and Economics by Glenn E. Handwerk and James H. Gary.

2. There is a large discrepancy between the CRI figure and the numbers derived here, but the difference is less than an order of magnitude. CRI's only note on their methodology reads—

Factors used: 5.78 MBtu/barrel crude oil. Source for average annual residential energy consumption (94.6 MBtu per household): U.S. Department of Energy, Energy Information Administration, Look at Residential Energy Consumption in 2001.

It is not clear what 5.78 Mbtu/barrel crude oil (equivalent) applies to. As noted, the boundary conditions used for these kinds of calculations vary from study to study.

3. I did not mean to pick on Dasani (Coca-Cola) in this article. I could have used Nestle, Pepsi, Evian or any of the others. Using one example made things easier.